Skip to content
Dashboard
Vercel Live

Nov 12: Intro to AI Gateway

Relace models

Relace is a hosted inference provider for open-weight AI models, delivering cost-efficient access optimized for coding, reasoning, and agentic workloads.
Model
Context
Input
Output
Latency
Throughput
Cache
Search
Capabilities
ZDR
No Training
Regional Inference
Free AI Gateway Credit
Released
1M
$0.24/M
$8.40/M
1.5 s93 tps
Read$0.20/M
—
+2
—10/23/2026
1M
$0.12/M
$0.50/M
1.0 s258 tps
Read$0.003/M
—
+2
—10/17/2026
1M
$0.12/M
$0.50/M
1.4 s—
Read$0.003/M
—
+1
—09/08/2026
1M
$0.04/M
$0.08/M
0.9 s94 tps
Read$0.03/M
—
+1
—08/26/2026
1M
$0.07/M
$4.38/M
1.9 s124 tps
Read$0.07/M
—
—08/18/2026
1M
$0.07/M
$0.14/M
0.6 s113 tps
Read$0.01/M
—
+1
—07/31/2026
1M
$0.18/M
$11.01/M
1.2 s105 tps
Read$0.18/M
—
+2
—07/16/2026